Sign in to follow this  
jnWake

Compressing/Limiting when Mastering

Recommended Posts

Hey,

When you are mastering a mix, how much do you usually allow your compressors/limiters to do in gain reduction?

I know you should "play" with the settings until you feel it sounds right, but I'm curious as to what are "common" levels of gain reduction when mastering. I'm always "afraid" of compressing too much so I always end up with pretty quiet mixes :razz:

Share this post


Link to post
Share on other sites

I have the same fear. Once I start seeing any gain reduction on the limiter my mind automatically assumes it's sounding bad and because i think it is, it does.

Just a slight reduction with no unintentional pumping is what i have been told to look for. If you hit that then back it off but i'm curious as to what other people think too.

Share this post


Link to post
Share on other sites

I use master limiter at -0.2 or 0db

Also max point to all tracks in the mix close to 0db.

Drums and guitars(rhythm) works at same position. 0db

After this i don't have to use lot of compression(-0.8 or -1.8db "IN" and 0db OUT) to the mix.

I don't have to use limiter/compressor if i cut freqs in "right direction"(depends of sound i want to get at the end).

Actually this question kinda specific imo. You should find your own way.

SFME

Share this post


Link to post
Share on other sites

I almost never compress the master channel more than 2:1, but that's for orchestral music. Other genres might need heavier compression.

Share this post


Link to post
Share on other sites

It does depend on the genre, but maybe about like 6 dB or so is good for general purpose reduction (assuming the master is usually around like -10 dB), then enough make-up gain to maximize. Use a compressor with a visual gain reduction meter to help train your ear. Endorphin is nice and also free.

Share this post


Link to post
Share on other sites

I generally try to shoot for an RMS or dynamic range level when limiting. TT dynamic range meter is a good tool for this purpose.

As for master bus compressors, I never use em. But people who do generally say you just want to be tickling the gain reduction. You'll want to mix into a bus compressor from the start, putting it on at the end kinda defeats the purpose.

Share this post


Link to post
Share on other sites

A compressor's gain reduction should visually move like an eye blinking.

If you slightly compress every track, bus, and master you should end up with a full sound.

If you need more compression on the master then you're mixing is probably off. If you want a more in your face sound from the master compressor, use 2 compressors in serial. Two compressors back to back that have a gain reduction of -3db will sound a helluva lot better than one compressor that has -6db. -6db is a little drastic IMO.

Also -12db on RMS in general is a good RMS to be at for most genres.

Share this post


Link to post
Share on other sites

I generally shoot for -6rms, but that really depends on the song. As far as gain reduction and compression, I do all that before I start mixing.

I set up my stereo bus chain before I start the mix, usually with a ratio of no more than 4:1 (usually 2:1) with a brickwall limiter on. I make up the gain via harmonic exciters instead of just driving the compression. This way I mix everything with regard to the compression, so I end up using much less compression on the individual tracks and get a cleaner/punchier/louder mix in the end. Also, doing things this way lets me fine tune the overall gain stage of the entire mix so I can get up as high as -2rms without heavy clipping (though I've never felt the need or have been asked to go that stupidly high).

The thing I go on about most of all is always headroom. I believe that the management of stereo bus headroom is the deciding factor in the punch:volume ratio of a mix. My philosophy is to control the low end so it doesn't eat up bandwidth and don't crowd the high end (everything doesn't need to shimmer) as to keep the master compressors from overworking in a negative way.

Share this post


Link to post
Share on other sites
If you want a more in your face sound from the master compressor, use 2 compressors in serial. Two compressors back to back that have a gain reduction of -3db will sound a helluva lot better than one compressor that has -6db. -6db is a little drastic IMO.

Nice, I don't think I've ever tried that and then taken it seriously before. I think once, I tried it and it worked to my ears, but I never thought much of it until now.

In typical tracks, I just use a soft knee limiter, compress my drums, and have the limiter peak at 0dB, plus a spectroscope for an amplitude check. If I want extra power, I put some parallel compression in. In those cases I tend to experiment with the sound of compressor after the limiter or limiter after the compressor, and limiter after the compressor tends to sound less punchy to me since the compressor I use ends up leaving the final amplitude kind of unchanged and at or under 0dB with more punchiness, rendering the limiter kind of useless.

Share this post


Link to post
Share on other sites

I just want to pop in and remind anybody who is looking for advice in this thread that there are many ways to skin a cat, and no one method is right or wrong. For instance, when Avaris says

If you need more compression on the master then you're[sic] mixing is probably off.
He does so with more authority than is really appropriate. Listen to everyone's ideas and try them out, but don't believe anyone who says their way is the right way.

Also, Snapple knows his stuff, but be aware that -6dBFS RMS is what is technically referred to as "Holy Shit Loud." That's Skrillex loud. It may not be what you're going for.

Share this post


Link to post
Share on other sites

I like to gently compress my drum bus, and gently compress my bass, if needed and if it adds punchiness to the sound (as desired to taste). On the master, I gently compress again, no more than 5db GR and only on the high peaks (snare, kick, usually snare), tiny bit (2db) makeup gain if needed. Then, soft-knee limiter, only a couple of db (2-4ish) GR and again *only* on the wild peaks, unless I'm looking for that smashed effect (typically NOT haha!). "Eyeblink" is a good descriptor for what I like to see happening in my master compressor and limiter.

-6db RMS holy shit dude. *covers ears* I was taught -12 is a good number but I usually end up in the -11 to -10 range, -9 tops and certainly not for the entire track!

Share this post


Link to post
Share on other sites

Also, Snapple knows his stuff, but be aware that -6dBFS RMS is what is technically referred to as "Holy Shit Loud." That's Skrillex loud. It may not be what you're going for.

Oh hell yeah, -6dB is ridiclously loud, but I like being loud. And that's another thing worth noting, that the elements in your song contributing to the loudness make all the difference. When I mix to that level I have very carefully EQed guitars being hard panned and all kinds of things that aren't going to sound as loud as they actually are. Stereo imaging also tricks your brain with volumes and levels, and harmonic exciters (maximizers) do a huge job of raising RMS without destroying the mix.

With careful EQ and compression I get mixes that are loud but never clip (even with all stereo bus effects off, including limiters). Going back to the issue of stereo bus headroom, no channel going into the stereo bus is peaking past -10dB, this gives my master compressors all the headroom they need to keep from breaking the sound.

Share this post


Link to post
Share on other sites

I don't really see anything in master processing that couldn't be accomplished more accurately/surgically during mixing without introducing artifacts on the whole package.

Share this post


Link to post
Share on other sites
I don't really see anything in master processing that couldn't be accomplished more accurately/surgically during mixing without introducing artifacts on the whole package.

Well, what we do in our DAWs in the "mastering" stage of a songs production is different from what actual mastering of a song/record is. Most of what mastering is cannot be done in the mixing stage alone, that's why the process is broken up. I think what we here refer to as mastering is the process of getting cohesion in the mix at the last stage of the chain.

And really, no amount of individual track mixing can do what compressing the entire stereo bus does, even if you're just barely compressing it.

But yeah, RMS is getting pretty unreliable for me these days, because what I hear doesn't always match what I see in number values. I've started metering in EBU and things are finally starting to translate correctly between the actual physical world and the digital world.

Share this post


Link to post
Share on other sites
Hooray for being a noob and not knowing most of what any of you are talking about!

First things first, Joe! At your rate of speed, you'll be here in a month or so. ;-)

Share this post


Link to post
Share on other sites
I don't really see anything in master processing that couldn't be accomplished more accurately/surgically during mixing without introducing artifacts on the whole package.

Master compression and limiting are ways of adding cohesion to all instruments and drums together while keeping everything in the same-ish volume range. If you have mixed well, you shouldn't need too much master compression to get cohesion. Limiting of course keeps the track from going over 0db at any point. (or less... -0.1 or so). Snappleman is right, you can't do this on the individual tracks only.

Keeping in mind... none of this should be considered actual "mastering."

Share this post


Link to post
Share on other sites

I usually bounce my mixes to a single audio file, and they're peaking generally at around -3 to -6dB. I then start a new session and add MASTER on the end.

I start off with Multi-band compression, with ratio's of 2:1 and varying attack and release times for each band. I use faster attack times for music with drums and stuff, and slower attack times for ambient music. Typically gain reduction is anywhere from 0.5dB to 2dB max.

An EQ is next in my chain where again, i usually dont cut or boost more than 1.5dB, and I use wide Q's. Subtly is the key here, since your mix should already be sounding great.

Finally, I have a limiter at the end with the limit set at -0.2dB. I increase the gain by slightly more than the loudest part of my entire mix.

This is Syllix's home brain surg... I mean, mastering method. :D

Share this post


Link to post
Share on other sites

You'll have to be more specific with what you mean with "cohesion", since the way I see it that's exactly what you'd accomplish by tweaking the individual signals. Same with making sure the volume never peaks above -0.1db.

To me, putting a compressor on the master is the equivalent to trying to make a picture larger by resizing it with bicubic or nearest neighbour. Actually, that's probably too generous of a comparison since with the former you're erasing information that you can't get back afterwards.

Share this post


Link to post
Share on other sites
You'll have to be more specific with what you mean with "cohesion", since the way I see it that's exactly what you'd accomplish by tweaking the individual signals. Same with making sure the volume never peaks above -0.1db.

To me, putting a compressor on the master is the equivalent to trying to make a picture larger by resizing it with bicubic or nearest neighbour. Actually, that's probably too generous of a comparison since with the former you're erasing information that you can't get back afterwards.

Think of it as compressing a single sample. Once the mix gets to the stereo bus it's basically a sample playing through it. Break it down to something like a snare drum, if you compress the snare drum itself, you get some characteristics out of it, where as compressing the entire song gets different characteristics out of the snare drum because of the context that it's in.

Compressing the entire mix at that stage also brings up things between the transients and kind of cooks the song together a little bit. If you compress too much then you get artifacts and squashed dynamics, and that's where the balance comes into it.

Share this post


Link to post
Share on other sites
Keeping in mind... none of this should be considered actual "mastering."

This kind of confuses me. What is actual mastering then?

Share this post


Link to post
Share on other sites
This kind of confuses me. What is actual mastering then?

Oooh ooh wait I think I know this one.

Mastering is actually an album term, when you bring every track on an album in line so that it all sounds alike in volume, timbre, etc. You kind of apply the same principle except you're treating almost all the songs on the album as a single track that you need to apply things to.

Yes? Do I win?

Share this post


Link to post
Share on other sites
Oooh ooh wait I think I know this one.

Mastering is actually an album term, when you bring every track on an album in line so that it all sounds alike in volume, timbre, etc. You kind of apply the same principle except you're treating almost all the songs on the album as a single track that you need to apply things to.

Yes? Do I win?

Yeah that's part of it. Mastering also includes very precise EQ, compression, imaging, limiting and sometimes even reverb. The big differences are that in mixing you're working on the balance to get the most out of the instruments, and make them fit together correctly within the musical context of the song, in mastering you're bringing out the most of the mix and trying to make it as sonically pleasing and even as you can. Another big part of mastering is getting the cues down, song lengths, fidelity, and formatting so you can deliver a package in whatever medium is required. The end result in mastering these days is usually a well rounded sounding disc image that can be used for production.

Share this post


Link to post
Share on other sites
Oooh ooh wait I think I know this one.

Mastering is actually an album term, when you bring every track on an album in line so that it all sounds alike in volume, timbre, etc. You kind of apply the same principle except you're treating almost all the songs on the album as a single track that you need to apply things to.

Yes? Do I win?

Yeah, that's pretty much it. A lot of people just think of mastering as "smash the crap out of the track".

Which, sadly, for the case of many "re-mastered" albums, is more or less what they did. :)

Share this post


Link to post
Share on other sites
Compressing the entire mix at that stage also brings up things between the transients and kind of cooks the song together a little bit. If you compress too much then you get artifacts and squashed dynamics, and that's where the balance comes into it.

Compression of any level (and EQ boosting for that matter) on the master also adds intermodulation distortion. This is basically inharmonic non-linear artifacts from a sum of different frequencies that is pretty much universally regarded as unpleasant (and AFAIK is scientifically linked with hearing damage and speaker damage) unlike harmonic distortion which is the desired type used for distortion processing. Subtle processing will result in subtle intermodulation distortion, but gets more apparent and eventually unbearable as you raise the volume. Of course, most modern music is so processed that you can barely turn the knob above zero before things start to feel grating.

If I'm feeling extra cynical I'll say that the whole mastering culture as we know it today came about when CDs were established as the standard format, and former vinyl mastering engineers had to think up a way to keep their jobs by claiming that the produced material needed to be processed further by them before ending up on CD.

And then we have the radio which further added to this development. Radio has a huge psychological authority on our society/culture. If a song is played on the radio, then it must obviously have professional merit if it ended up there for all to hear.

Multi-band compressors were not born as commercial products for production use. They were originally McGyvered by radio engineers competing with eachother to make their stations feel louder. Now, if there's one thing I've learned about trends it's that they have no correlation to quality (kinda like how people percieve 24fps as better than higher framerates because it feels more cinematic). So once people got used to the compressed material broadcasted from radio stations, musicians started wanting that sound on their CDs, since if it sounded more like radio then obviously that meant it sounded more professional. And from there things started snowballing out of control.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this